We propose to plan a research project on diabetes mellitus among the Metlakatla Indian Community of Annette Island, Alaska. About 1,525 subjects reside in this community, of whom about 1,100 are Tsimshian Indians. A preliminary study performed by us of clinical diabetes mellitus prevalence across the entire state of Alaska revealed that the highest prevalence occurred among Alaska Native residents of Annette Island. Twenty to thirty years ago, however, diabetes rarely occurred among Alaska Natives. Over this time period, lifestyle has changed considerably with the increasingly widespread availability of store bought food and labor saving machines. Preliminary goals of the research project will include measurement of diabetes mellitus prevalence, ascertainment of risk factors for diabetes mellitus, and the design of a behavioral intervention to decrease the prevalence of this metabolic disorder. The specific objectives and procedures to be followed to achieve them will depend on what is deemed feasible and of interest to the Native community and investigators during the planning process. At the current time, the following tentative plan for the research project is as follows. Oral glucose tolerance testing of all Alaska Native residents of Annette Island age 20 years and older will be conducted. Due to the youthful age structure of the population, about half of Annette Island residents will be included in this survey. World Health Organization criteria will be used to assign the diagnosis of normal glucose tolerance, impaired glucose tolerance, and diabetes mellitus. We expect all diabetes found to be NIDDM, due to the rarity of IDDM among Alaska Natives. Fasting insulin and C-peptide levels will be measured to assess the prevalence of insulin resistance among subjects with normal glucose tolerance. Instruments to assess dietary intake and physical activity will be developed with the help of collaborators. These lifestyle variables will be compared in subjects with normal and abnormal glucose tolerance to assess their effect on diabetes prevalence.